Author's personal copy

Plant Systematics and Evolution (2019) 305:255–268 https://doi.org/10.1007/s00606-019-1565-0

ORIGINAL ARTICLE

The Struthiopteris spicant (Blechnaceae, Polypodiopsida) complex in Western Europe, with proposals for taxonomic and nomenclatural changes

Sonia Molino1 · Jose M. Gabriel y Galán1 · Pawel Wasowicz2 · Pablo de la Fuente1 · Emily B. Sessa3

Received: 7 August 2018 / Accepted: 9 January 2019 / Published online: 16 February 2019 © Springer-Verlag GmbH Austria, part of Springer Nature 2019

Abstract Struthiopteris spicant is a medium-sized, dimorphic fern that occurs mainly in Western Europe and Western North America. The species is quite variable, with several forms differing in frond sizes and degree of dimorphism. In addition to the typi- cal plant, in Europe two other varieties are recognized: the Iberian S. s. var. homophyllum includes plants up to 20 cm, monomorphic or subdimorphic, with fragmented cenosori; the Icelandic S. s. var. fallax comprises very small plants up to 5 cm, monomorphic, with isolated sori. Outstanding questions remain about hybridization among the different forms, their taxonomic status, and the relations between different populations. The present work aims to study additional morphological and anatomical features of the S. spicant complex in Western Europe using a statistical approach, to resolve the taxonomic position of the different forms. We observed traits ranging from the rhizome scales to the spores and used one-way ANOVA to test for significant differences between all varieties and PCA to determine whether plants assigned to different varieties can be classified statistically into distinct groups. We detected significant differences in several of the analysed traits among the varieties in the complex. Qualitative differences in pinna anatomy and epidermal cells also help segregation of the varieties considered. Our results allowed us to propose a new specific status for the Icelandic endemic var. fallax and a new endemic Spanish variety, S. s. var. pradae.

Keywords Endemism · Morphology · Pteridophytes · Taxonomy

Introduction

Handling Editor: Ricarda Riina. Blechnaceae Newman is a subcosmopolitan family of lepto- * Jose M. Gabriel y Galán sporangiate ferns (Polypodiopsida) that includes around 250 [email protected] species (PPG1 2016). Until recently, most of this diversity Sonia Molino fell into one large genus, Blechnum L., but after evidence [email protected] accumulated concerning its non-monophyletic status (Shep- Pawel Wasowicz herd et al. 2007; Gabriel y Galán et al. 2013; Perrie et al. [email protected] 2014; Gasper et al. 2017), Blechnum was split into sev- Pablo de la Fuente eral entities (Gasper et al. 2016). In its current conception, [email protected] Blechnaceae as a whole is made up of 24 genera (PPG1 Emily B. Sessa 2016). emilysessa@ufl.edu The genus Struthiopteris Scop., as currently defined, was resurrected by Gasper et al. (2016) to accommodate small to 1 Department of Biodiversity, Ecology and Evolution, Faculty of Biology, Universidad Complutense, Avenida Jose Antonio medium size, dimorphic, pinnate species (Fig. 1). It contains Nováis, 12, 28040 Madrid, Spain five species, with a northern geographical distribution. Four 2 Icelandic Institute of Natural History, Borgir vid Nordurslod, of these species are restricted to eastern Asia, in Japan and 600 Akureyri, Iceland Taiwan: Struthiopteris niponica (Kunze) Nakai, S. hancockii 3 S. amabilis S. cas- Department of Biology, University of Florida, Box 118525, (Hance) Tagawa, (Makino) Ching, and Gainesville, FL 32611, USA tanea (Makino) Nakai (Iwatsuki 1992; Chiou et al. 1994).

Vol.:(0123456789)1 3 Author's personal copy

256 S. Molino et al.

Fig. 1 Forms of plants within the Struthiopteris spicant complex. a var. homophyllum (MACB 1092626); d subdimorphic: S. spicant var. Dimorphic: S. spicant var. spicant (MA 747921); b monomorphic: homophyllum (MACB 32367). Bar = 2 cm in a; 1 cm in b; 2.5 cm in S. spicant var. fallax (MACB 109359); c monomorphic: S. spicant c, d

The fifth species, S. spicant (L.) Weiss, is the object of the sterile fronds have expanded pinnae without sporangia while present study. This taxon is described as a terrestrial or saxi- the fertile ones are much longer and have highly contracted colous plant, living on acid or neutral soils, with dimorphic pinnae without green tissue at maturity (Fig. 1a); b) mono- fronds: the sterile ones are lanceolate, pinnatisect, and form- morphic, in which all fronds are similar in morphology, ing a basal rosette of 20–70 cm long, and the fertile ones irrespective of the presence of sporangia (Fig. 1b, c); or c) are erect and longer, with heavily contracted pinnae which subdimorphic, in which some fronds, regardless of whether bear costal cenosori and have continuous indusia (Rolleri they bear sporangia or not, are equal or slightly longer than and Prada 2006). This species occupies a larger distribu- the rest, with slightly contracted pinnae (i.e. when fully tion than its congeners, occurring in almost all of Europe developed, they are, at least partially, narrower but with a (from Sweden, Poland, Ukraine, and Romania westwards clear area of green tissue) (Fig. 1d). to Iceland and the Iberian Peninsula), some parts of North- This variation in frond morphology in Struthiopteris spi- ern Africa (where it is rare), the Macaronesian archipela- cant has been the subject of previous research (e.g. Löve gos (Canary Islands, Madeira, and Azores Islands) and the and Löve 1966). Taxonomists have described several entities Pacific Northwest area of North America (USA: California, that have been formally recognized with different taxonomic Oregon, Washington, Alaska; Canada: British Columbia). It assignments, usually as subspecies or, more frequently, as is typically a very common element in a variety of habitats, varieties or ecological forms. In consequence, S. spicant particularly in forests (Lawalrée 1964; Nauman 1993). can be seen as a morphological complex. While most of Such a wide range has allowed a certain amount of mor- its forms have ultimately been considered to fall under the phological variation to emerge, which is mainly manifested expected variation of S. spicant s.l. and thus have been as differences in frond sizes and degree of dimorphism. neglected in local floras, some of them are widely accepted Thus, plants can be: a) dimorphic, in which there are two and regularly included in floras and taxonomic treatments. highly different types of fronds associated with function: the For example, the Pacific American plants fit the general

1 3 Author's personal copy

The complex Struthiopteris spicant in Western Europe 257 dimorphic morphology of S. spicant but with larger sterile genetic variation in chloroplast, nuclear, and microsatel- fronds, up to 100 cm (Löve and Löve 1968). Variety homo- lite data between European plants from many populations phyllum (Merino) Gabriel y Galán & R. Pino (Wasowicz and between these and American populations (unpublished et al. 2017b) comprises smaller plants, with erect fronds up data). This observation coincides partially with previous to 20 cm, usually all sporogenous, monomorphic or subdi- research that also included different populations of S. spicant morphic, and with the cenosori typically fragmented and (Soltis and Soltis 1988; Korpelainen and Pietiläinen 2008). the indusia discontinuous (Merino 1898; Ormonde 1986). The scarcity of molecular differences prevents construction This variety is endemic to the northwest of the Iberian Pen- of a well supported and highly resolved phylogeny that could insula, in both Spain and Portugal (Ormonde 1986; Molino help to elucidate the taxonomy. Here, we focus primarily on et al. 2018). Finally, var. fallax (Lange) Wasowicz & Gabriel the European taxa because the variation within the North y Galán comprises very small plants, only about 2–5 cm, American group is still poorly known and the latter would usually with monomorphic fronds, uncontracted pinnae, need its own study using a comprehensive sampling. The and isolated sori. It grows exclusively in Iceland, where its present work aims to study additional morphological and scarce populations live in close proximity to (and presum- anatomical features of plants within the S. spicant complex ably are dependent on) hot springs (Wasowicz et al. 2017b). from Western Europe, in order to resolve the taxonomic The consideration of these distinct forms as varieties position of the different forms. dates back to pteridologists Lange (1880) and Christ (1904). More recently, Löve and Löve (1968) observed that: a) cul- tured plants of homophyllum and fallax maintain their traits over time, and var. homophyllum is more variable than var. Materials and methods fallax; b) both have the same chromosome number, 2n = 68, which was later corroborated by Horjales et al. (1990) for A total of 50 individuals from all the varieties of the com- var. homophyllum; and c) both are completely interfertile plex were studied (Appendix 1). Plants came either from with the typical form of S. spicant, so it is possible that herbaria or from new collections (Fig. 2). At least one hybrids may form if they come into contact. Based on voucher per sampled population has been deposited in the these ideas, Löve and Löve proposed a taxonomy in which MACB herbarium (Madrid, Biology UCM). Samples were S. spicant (sub. Blechnum spicant) was divided into two taken to represent variation in the complex throughout the subspecies: a) subsp. nipponicum (Kunze) Löve & Löve Atlantic region of Europe, including the latitudinal extremes which includes the Pacific populations from both Japan and of its distributional range: Iceland and the Canary Islands. America (a concept which has been usually disregarded by More in-depth sampling was completed in the northeast of subsequent pteridologists (e.g. Nakato 1987); and b) subsp. the Iberian Peninsula where a higher incidence of morpho- spicant, the European race, with two varieties in addition to logical variation has been detected (personal observation of the typical var. fallax Lange and var. homophyllum Merino. the authors). A few individuals from southern France in the These authors never demonstrated conclusively the existence Mediterranean and the Alps in Austria were also included. of hybrids, but did state (op. cit., p. 670) about var. homo- The following traits were observed for all specimens: phyllum that “at least one of the specimens we cultivated size of rhizome scales (at least 3 per individual), length of could perhaps be interpreted as a hybrid”, and about var. sterile and fertile leaves, differentiating between petiole and fallax that “hybrids have not been observed for the simple laminae (at least 2 sterile and 1 fertile per individual, if pos- reason that the two populations never meet (…) but there sible), width at midpoint of sterile and fertile pinnae (at least is no reason to believe that such hybridization would not 1 per frond) and anatomical characters of fertile and sterile be possible if an opportunity arises”. In view of the above, pinna sections (2 per taxon), and the epidermis, including S. spicant can be described as a complex of forms not fully both adaxial and abaxial surfaces (at least two samples per understood to date. Outstanding questions remain about taxon). In addition, we measured the width and length of hybridization among the different forms, their taxonomic 20–30 spores in at least two individuals per taxon. Figure 3 status, their geographical distributions, and the relations of shows schematically the traits observed. the European and American populations to one another and To characterize the level of sterile–fertile leaf dimor- to the Asian species. phism, we also calculated two ratios: one is the result of In the context of a broader study focused on the whole dividing the length of the fertile lamina by the length of its genus Struthiopteris, we have undertaken a new approach petiole (since dimorphic plants tend to elongate the fertile to the S. spicant complex in Europe by applying DNA petiole), and the other is the result of dividing the width sequencing and observing new morphological features not of the fertile pinna by the width of the sterile pinna (since considered by previous researchers. Regarding DNA char- dimorphic plants tend to reduce the width of the fertile acters, a noticeable and striking result is the very limited pinnae).

1 3 Author's personal copy

258 S. Molino et al.

Fig. 2 Atlantic region of Europe, showing the sampling locations. Islands (Spain) in the south. The inset map indicates with more detail This sampling covers the latitudinal extremes of the distribution area the sampling locations in the Iberian Peninsula (Spain and Portugal) of Struthiopteris spicant, from Iceland in the north to the Canary where a higher degree of morphological variation has been detected

The following traits were measured using a light micro- post hoc Tukey test to test for significant differences between scope with Nikon Coolpix camera attached: 1) length of the investigated morphological forms in the S. spicant complex. sterile laminae, 2) width of the sterile laminae, 3) length of Principal components analysis was performed on the matrix the sterile petiole, 4) width of the sterile pinnae, 5) length of containing the data from measurements of the fronds and the fertile laminae, 6) width of the fertile laminae, 7) length rhizome scales in order to explore the morphospace. Signifi- of the fertile petiole, 9) width of the fertile pinnae, 11) length cance level was set to α = 0.05 for all the tests. All statistical of the rhizome scale, 12) width of the rhizome scale, 13) analyses were done with the software IBM SPSS Statistics length of the spore, and 14) width of the spore. Key ratios 25 (IBM). between some of these variables were subsequently calcu- lated: 8) ratio of fertile laminae length/fertile petiole length, and 10) ratio of sterile pinnae width/fertile pinnae width. When several measurements of a particular trait were Results obtained from the same individual, mean values and stand- ard deviations are reported. Morphoanatomical features

Statistical analyses The measured traits fall broadly into three classes of char- acters related to lamina, rhizome, and spore morphology. We calculated basic statistics (i.e. means and standard devia- We detected significant differences in several of these tions) for all measured traits and used one-way ANOVA and traits among the four varieties of the Struthiopteris spicant

1 3 Author's personal copy

The complex Struthiopteris spicant in Western Europe 259

with the middle part almost not sclerosed, entire margins, and flagelliform apex. Struthiopteris spicant var. fallax has scales similar to those of var. homophyllum, that are lanceo- late, concolorous (pale brown), without a sclerosed middle section, entire margins, and flagelliform apex. Regarding the frond characters, we can distinguish two groups of taxa, one composed of S. spicant var. fallax and var. homophyllum, which have significantly shorter fronds, and the other by S. spicant var. spicant and var. pradae, which have significantly larger fronds (Tables 1, 2). In the case of some variables, such as sterile and fertile lamina width and sterile pinna width, the group homophyllum + fal- lax splits into two further, statistically significant groups, with fallax having the smaller dimensions. The trait “length of fertile laminae” differs significantly among all four taxa with var. pradae having the largest laminae, followed by var. spicant, var. homophyllum, and finally var. fallax. The ratios calculated for the lamina dimensions also differ between taxa. The fertile lamina/petiole length ratio forms two groups, one with S. spicant var. homophyllum and var. fallax, which do not elongate the petioles in the fertile fronds, and the other group with var. spicant and var. pradae, which do so. The width of sterile pinnae/fertile pinnae ratio also formed two significantly different taxon groups, one with S. spicant var. fallax (monomorphic), and the other with var. spicant and var. pradae (dimorphic). Struthiopteris spi- Fig. 3 Schematic of the traits observed. a Macromorphological vari- cant var. homophyllum overlaps both groups (subdimorphic). ables regarding general sizes of both types of leaves. b Micromor- Struthiopteris spicant var. pradae has a second type of phological variables regarding sizes of pinnae. c Micromorphological sporophyll with the same general appearance as the tropho- variables regarding sizes of rhizome scales. Refer to Table 1 for more phyll. We measured 10 of these fronds, but did not include detailed information on the names of the variables. Bar = 5 cm in a; 375 µm in b; 800 µm in c their measurements in the table since there is no equiva- lent frond type in the other varieties. The results for those fronds were laminae 30.14 (± 13.33) cm long, 4.36 (± 1.29) complex (Table 1). One of these varieties, var. pradae, is cm wide, petioles 10.66 (± 9.46) cm long, and pinnae 0.31 newly recognized here on the basis of these data (see below). (± 0.09) cm wide. For characters related to rhizome scales, the width of the The spore length character again discerned two groups base differed significantly between var. fallax and var. pra- of taxa, one with shorter spores formed by S. spicant var. dae, while var. homophyllum and var. spicant overlapped homophyllum and var. fallax, and another with larger spores with each other and one of the other two varieties (Table 1). formed by S. spicant var. spicant and var. pradae. For spore The length of the rhizome scales differed significantly width, there is a group formed by var. homophyllum and var. between three groups: S. spicant var. fallax had the shortest fallax, with narrower spores, another one with var. spicant scales, var. homophyllum had scales of intermediate length, and wider spores, and var. pradae overlapping both groups. and var. spicant and var. pradae had the longest scales. We carried out a principal components analysis on a Besides these quantitative differences, qualitative differ- data matrix containing the data from measurements of the ences exist as well (Fig. 4, Table 2). Struthiopteris spicant fronds and rhizome scales. Factor loadings for each variable var. spicant has bicolorous (dark brown and pale brown), in each component are shown in Table 2. Three components linear-lanceolate scales, with the middle section narrowly explained 75.67% of the variance among taxa (Fig. 5). The sclerosed from the base to the apex, entire margins, and PCA results (Fig. 5a, b) show that two major groups can be flagelliform apex. Sometimes simple, pluricelular, uniseri- identified: one formed by three overlapping taxa, S. spicant ate, and glandular hairs can be found on the margins of the var. spicant, var. pradae, and var. homophyllum, and another scale. Struthiopteris spicant var. pradae has scales with the group formed by S. spicant var. fallax, which is clearly dif- same characteristics as var. spicant. In contrast, var. homo- ferent from the rest of the taxa studied. When only com- phyllum has concolorous (pale brown), triangular scales, ponents 2 and 3 are considered (Fig. 5c), there is a general

1 3 Author's personal copy

260 S. Molino et al.

Table 1 Morphological traits measured in the four varieties in parentheses. Taxa with the same letter after the parenthesis for a within the complex Struthiopteris spicant. Values are expressed as particular variable show no statistical differences, as detected by a mean ± standard deviation, followed by number of samples measured post hoc Tukey ANOVA (α = 0.05)

S. spicant var. fallax S. spicant var. homophyllum S. spicant var. spicant S. spicant var. pradae Traits relating to laminae/fronds st_lam_length [cm] 1.55 ± 0.31 (4)a 7.49 ± 1.42 (13)a 21.46 ± 6.18 (67)b 26.56 ± 10.98 (25)b st_lam_width [cm] 0.5 ± 0.08 (4)a 1.62 ± 0.86 (13)b 3.12 ± 0.93 (67)c 3.52 ± 1.24 (25)c st_pet_length [cm] 0.23 ± 0.13 (4)a 1.15 ± 0.86 (13)a 4.93 ± 2.88 (67)b 6.17 ± 5.84 (25)b st_pin_width [cm] 0.16 ± 0.05 (4)a 0.29 ± 0.09 (13)b 0.37 ± 0.73 (67)bc 0.39 ± 0.09 (25)c fer_lam_length [cm] 2.8 ± 1.36 (4)a 11.26 ± 2.85 (24)b 28.35 ± 9.82 (40)c 39.01 ± 13.6 (8)d fer_lam_width [cm] 0.75 ± 0.16 (4)a 2.12 ± 0.55 (24)b 4.04 ± 1.35 (40)c 4.81 ± 1.07 (8)c fer_pet_length [cm] 0.45 ± 0.34 (4)a 2.47 ± 1.71 (24)a 10.91 ± 4.77 (40)b 14.58 ± 7.95 (8)b fer_lam/pet_length 0,16 ± 0.09 (4)a 0.21 ± 0.55 (24)a 0.39 ± 0.17 (40)b 0.37 ± 0.15 (8)b fer_pin_width [cm] 0.17 ± 0.08 (4)a 0.2 ± 0.05 (24)a 0.17 ± 0.04 (40)a 0.18 ± 0.05 (8)a st_pin/fer_pin 1,04 ± 0,67 (4)a 0.78 ± 0,7 (13)ab 0.42 ± 0.16 (40)b 0.4 ± 0.15 (8)b Traits relating to rhizomes rh_sc_length [µm] 1705 ± 272.03 (6)a 3609.72 ± 980.39 (27)b 5581.14 ± 2029.11 (57)c 6391.67 ± 1236.93 (30)c rh_sc_width [µm] 408.33 ± 148,88 (6)a 737.0.4 ± 200.4 (27)ab 1673.68 ± 454.34 (57)ab 971.25 ± 316.89 (30)b Traits relating to spores spore_length [µm] 43,33 ± 7,05 (60)a 37,93 ± 4,79 (98)b 40,54 ± 3,27 (59)c 42,33 ± 5,24(150)c spore_width [µm] 31,83 ± 6,24 (60)a 26,56 ± 4,45 (98)b 28,61 ± 3,63 (59)c 29,43 ± 3,31 (150)ac

Key to variable names: st_lam_length length of the sterile laminae, st_lam_width width of the sterile laminae, st_pet_length length of the ster- ile petiole, st_pin_width width of the sterile pinnae, fer_lam_length length of the fertile laminae, fer_lam_width width of the fertile laminae, fer_pet_length length of the fertile petiole, fer_lam/pet_length ratio of fertile laminae length/fertile petiole length, fer_pin_width width of the fertile pinnae, st_pin/fer_pin ratio of sterile pinnae width/fertile pinnae width, rh_sc_length length of the rhizome scale, rh_sc_width width of the rhizome scale, spore_length length of the spore, and spore_width width of the spore

Fig. 4 Rhizome scales in taxa of the Struthiopteris spicant complex: a S. spicant var. fallax, b S. spicant var. spicant, c S. spicant var. pradae, d S. spicant var. homophyllum. Bar = 500 μm overlap of all taxa. This can be explained by the width of the which are contracted compared to the sterile pinnae, and rhizome scales and width of fertile pinnae, and variables that with the margins slightly curved towards the abaxial side. do not differ significantly between groups in the ANOVA. The indusia are complex and thick (Fig. 6a). Struthiopteris Regarding anatomical sections of the pinnae (Fig. 6, spicant var. fallax is a monomorphic plant, so its fertile Table 3), S. spicant var. spicant possesses fertile pinnae pinnae have similar anatomical features to the sterile ones,

1 3 Author's personal copy

The complex Struthiopteris spicant in Western Europe 261

Table 2 Morphological variables of the complex Struthiopteris spi- and indeed most fronds are fertile (Fig. 6b). Struthiopteris cant used in the PCA with the factor loadings for each component. spicant var. pradae has the same characteristics as var. spi- For variable explanations see Table 1 cant for its standard sporophyll, but in addition it possesses Component a second type of sporophyll which is straighter, more like 1 2 3 the trophophyll, and with a shorter indusium compared to the fully fertile pinnae (Fig. 6c). Struthiopteris spicant st_lam_length 0.155 0.030 0.108 var. homophyllum has shorter indusia, and its fertile pin- st_pet_length 0.129 0.031 0.071 nae are similar than the sterile ones, due to their subdimor- st_lam_width 0.151 − 0.017 0.010 phic condition (Fig. 6d). In S. s. var. spicant, the pinnae st_pin_width 0.119 − 0.130 − 0.162 are not curved in section, so the margins appear almost fer_lam_length 0.143 0.058 0.290 aligned with the meristele (Fig. 6a), whereas the margins are fer_pet_length 0.153 0.004 0.043 slightly curved in var. homophyllum (Fig. 6d) and var. pra- st_pet_length 0.135 0.096 0.224 dae (Fig. 6c) (i.e. the margins are slightly displaced towards fer_pin_width − 0.006 0.428 0.201 the abaxial surface from the plane of the meristele). Pinnae fer_lam/pet_length 0.097 − 0.078 − 0.257 appear strongly curved in the case of S. spicant var. fal- st_pin/fer_pin − 0.075 0.376 0.285 lax (Fig. 6b), with the margins very displaced towards the rh_sc_width − 0.004 − 0.317 0.637 abaxial surface from the plane of the meristele. rh_sc_length 0.109 0.225 − 0.334 The epidermis in all taxa shares the following features: the stomata are arranged following the veins, and simple, plurice- lular, uniseriate and glandular hairs are present on both abaxial and adaxial surfaces, as well as on the rachis. Some differences

Fig. 5 Dispersal graphics of the three components resulted from the PCA labelled by taxon. a Component 1 versus component 2; b Component 1 versus component 3; c Component 2 versus component 3

1 3 Author's personal copy

262 S. Molino et al.

Fig. 6 Anatomical sections of pinnae in taxa of the Struthiop- teris spicant complex. The left column shows photographs of sections of fertile pinnae; the black lines pass through the middle of the meristele in each case and serve as a reference for the degree of abaxial displace- ment of the pinna margins. The right column shows schematic drawings comparing the width of sterile pinnae (above) and fertile pinnae (below) for each taxon. a Strongly dimorphic var. spicant, with margins not or only slightly displaced. b Monomorphic S. spicant var. fallax, with margins strongly displaced. c Strongly dimorphic var. pradae, with two types of fertile pinnae: uncontracted (right column middle drawing) and contracted (right column lower drawing); margins are somewhat displaced. d Subdi- morphic var. homophyllum with margins somewhat displaced. Bar = 250 μm in photographs, 450 μm in schematic drawings

Table 3 Summary of the main morphological and anatomical features distinguishing the different taxa in the Struthiopteris spicant complex S. spicant var. spicant S. spicant var. homophyllum S. spicant var. pradae S. spicant var. fallax

Rhizome scales Linear-lanceolate, bicolorous Triangular, concolorous Linear-lanceolate, bicolorous Lanceolate, concolorous Frond dimorphism Dimorphic Subdimorphic Dimorphic (with two types of Monomorphic sporophylls) Laminae length Normally more than 20 cm No more than 20 cm Normally more than 20 cm No more than 5 cm Epidermis Anticlinal walls strongly Anticlinal walls strongly Anticlinal walls strongly Anticlinal walls slightly undulated undulated undulated undulated

were also observed in the epidermis cells and stomatal pat- Proposal of a new status for Struthiopteris spicant tern (Fig. 7) as follows: in both the abaxial and the adaxial var. fallax surfaces, the epidermis of S. spicant var. fallax possesses rec- tangular cells with anticlinal walls that are slightly undulated, Struthiopteris spicant var. fallax is a unique plant. Its gen- and rounded stomata with narrower companion cells that are eral morphology makes it highly probable that it belongs sometimes shared by more than one stomata. The remaining to the S. spicant complex, but several characters set it taxa share an epidermal pattern that differs notably from S. apart, including: a) traits regarding sterile reproductive spicant var. fallax. The abaxial and adaxial epidermis have structures are quite different from those in var. spicant elongated cells with highly undulated anticlinal walls and (e.g. size and morphology of the rhizome scales, overall ellipsoidal stomata with the same disposition as in (but larger size of the plant, monomorphic fronds, anatomy of the than) S. spicant var. fallax. pinnae and epidermis, and size of the spores); b) it has

1 3 Author's personal copy

The complex Struthiopteris spicant in Western Europe 263

Fig. 7 Epidermal patterns in abaxial (a–d) and adaxial (e–g) leaf surfaces of members of the Struthiopteris spicant complex: a S. spicant var. spicant; b, e S. spicant var. homophyllum; c, f S. spicant var. pradae; d, g S. spicant var. fallax. Bar = 80 μm in a–d, 100 μm in e–g an isolated, highly localized distribution, occurring only have never been detected, contrary to what occurs between near hot springs in Iceland; and c) it has a very particu- var. spicant and the rest of the varieties considered in this lar ecology, related to its geothermal environment. Thus, work. In consequence, it seems to us that var. fallax can although var. spicant also occurs in Iceland, populations of be better seen as a geographically and ecologically iso- these two entities are ecologically well differentiated and lated lineage, which falls completely within the concept their niches never overlap. As a result, intermediate, poten- of a unique species. The formal taxonomic treatment is tially hybrid plants between var. spicant and var. fallax presented below.

1 3 Author's personal copy

264 S. Molino et al.

Fig. 8 Holotype of Struthiopt- eris spicant var. pradae, MACB 110654. a Individual showing the typical three types of fronds. b Detail of the sporogenous area of the short fertile frond, with plain, uncontracted pinnae and interrupted sori. c Rhizome scale. Bar = 2 cm in a, 0.7 cm in b, 650 μm in c

Description of a new variety in Struthiopteris spicant Plants with this unique three-frond morphology have been located to date in four different and distant localities Plants with sterile fronds and two distinct types of fertile (Fig. 9). One has been visited continuously for 3 years, and fronds were found in several localities in the north of the new plants with this form were always detected, ranging Iberian Peninsula. Young individuals bear several fronds from small, young individuals to large, older ones. that resemble typical “sterile spicant” fronds, most of The general appearance and almost all the morphological which are sterile but one or two of which are at least par- details undoubtedly bring this plant into S. spicant. How- tially sporogenous, non-contracted, and with interrupted ever, the constant presence of the characteristic third type sori and indusia. Mature individuals develop, in addition to of partially sporogenic leaf with flat pinnae in individuals these two frond types, at least one fully sporogenous frond from populations that are distant from one another and per- that resembles the typical “fertile spicant” frond, with sistent in time suggests that these populations represent a highly contracted pinnae and continuous cenosori (Fig. 8). race that can be considered as different from the typical form Spores from both types of sporophylls appear morphologi- of S. spicant. In consequence, we proceed here to name and cally sound and fertile. describe this race as a new variety. The formal taxonomic treatment is presented below.

1 3 Author's personal copy

The complex Struthiopteris spicant in Western Europe 265

Spores are another extremely important character for clas- sifications in ferns (Tryon and Lugardon 1991). The spores of S. spicant var. spicant have been described in several studies, and those previous descriptions agree well with the characteristics found in this work (Lugardon 1965, 1974; Tryon and Lugardon 1991; Rolleri and Prada 2006; Pas- sarelli 2007; Passarelli et al. 2010). Ours are the first data reported for the remaining taxa, and we find significant dif- ferences in spore size between several taxa (Table 1). Since differences in size have been found for S. spicant var. homo- phyllum and S. fallax compared to S. spicant var. spicant, it would be interesting to carry out a more detailed study of additional characters of perispore morphology to determine whether further traits exist that would be useful to discrimi- nate these entities. Cross sections of pinnae have also proved very informa- tive and offer valuable characters to separate genera and even species in the family Blechnaceae (Prada et al. 2016; Fig. 9 Distribution map of Struthiopteris spicant var. pradae, show- Vicent 2017). Previous studies have described the anatomy ing the four distant known localities in northern Spain. Bar = 100 km and associated structures to the fertile pinnae for S. spicant var. spicant, which agree with our observations here. We also present the first pinnae cross sections for the remain- Discussion ing taxa in the complex and show their utility to distinguish them (Table 1, Fig. 6). While differences in the shape and size of fronds have previ- The morphology of the epidermis is another character ously been reported in S. spicant var. homophyllum and S. newly described here for all the entities in this study except fallax (Merino 1898; Löve and Löve 1966; Ormonde 1986; for S. spicant var. spicant, which has been already described Wasowicz et al. 2017b), in the current work we present the (Rolleri and Prada 2006) and whose main characteristics first statistical support for such differences based on ANOVA agree with the data reported in this study. The chief differ- and principal components analyses that distinguish clear ences we identified pertain primarily to S. fallax, which has groups within the taxa studied. It is clear from our data and cell walls only slightly undulating compared to the remain- analyses that the general morphology and the size of the ing taxa we considered (Table 2, Fig. 7), and which consti- fronds are crucial characters to distinguish S. fallax and all tutes one piece of evidence supporting its separation from the taxa within the S. spicant complex. In the case of S. fal- the S. spicant complex. lax, its size, degree of dimorphism, and isolated location Regarding the new status of S. fallax, even though its gen- support its elevation to the species level, and in the case eral morphology and chromosome number (Löve and Löve of the new variety S. spicant var. pradae, the presence of a 1968) would equally well support that this taxon is part of unique second type of sporophyll sets it apart. the S. spicant complex, we have also considered the follow- Rhizome scales are structures widely used by fern tax- ing facts: a) we reported in this study a number of charac- onomists and are particularly important in the Blechnaceae ters that are significantly different from S. spicant (including (Kramer et al. 1990; Rolleri and Prada 2006; Rothfels et al. general macromorphology, rhizome scales, pinnae anatomy, 2012) to discriminate genera and species. In this study, the and cells of the epidermis); b) there is a hard ecological rhizome scales observed in S. spicant var. spicant agree boundary between the two taxa and as a result their niches with a previous work (Rolleri and Prada 2006) except for and local habitat never overlap (in Iceland S. spicant is most the description of the margins, which are described as den- common in snowbeds and ravines, while S. fallax is confined tate in that study but entire in this work. Rolleri and Prada to high-temperature geothermal areas); c) the presence of (2006) studied just three individuals, while we have exam- intermediate individuals between S. spicant and S. fallax has ined more than 30, so the discrepancy in our observations never been confirmed (Wasowicz et al. 2017b); and d) when may be due to differences in sampling. In the rest of the taxa, cultivated far from its natural distribution area, S. fallax pre- the rhizome scales have not been described before, and we serves the same morphological characteristics and does not demonstrate that they are useful for distinguishing these taxa begin to resemble S. spicant (Löve and Löve 1968). For all in the case of S. spicant var. homophyllum, var. spicant, and of these reasons, we raise this taxon to the level of species S. fallax (Table 1).

1 3 Author's personal copy

266 S. Molino et al. and find it plausible that it could be a close relative of var. fronds could be a gradual process, and that the manifesta- spicant that has become adapted to hot springs. tions of dimorphism is transitional: thus, for many ferns, We also propose a formal status for S. spicant var. pradae, “many fertile fronds are merely “subdimorphic” in the based mainly on the presence of two type of sporogenous sense that the changes displayed are incomplete in kinds fronds in addition to the normal sterile ones, as noted above, and amounts”. This describes the case of plants like S. a distinct morphology that merits taxonomic recognition. spicant var. pradae well. This taxon and others previ- The presence of two distinct types of fertile fronds could ously documented, such as some members of the polypo- led to the consideration that the plant is trimorphic, but this diaceous genus Pyrrosia Mirb. (Hovenkamp 1986), are is not necessarily the case and deserves some explanation. essentially different from both dimorphic and trimorphic Pteridologists describe a fern as dimorphic when there is a ferns. Strictly speaking, we cannot consider this a case distinction between sporophylls and trophophylls (Wagner of true dimorphism or trimorphism, but a case somewhat and Wagner 1977). The term is often used somewhat infor- intermediate between the two. This striking fact deserves mally to acknowledge this characteristic, because it is recog- further investigation to elucidate the status of such “pseu- nized that dimorphism “may sometimes apply to other than dotrimorphic” ferns. fertile-sterile (i.e., various types of heterophylly of vegetative leaves in ferns)…”, and is somewhat arbitrary “because of so many transitions that occur between monomorphy and Taxonomic treatment dimorphy” (op. cit., p. 251). True dimorphism in leaves is a common feature in Identifcation key for the European taxa ferns, and occurs in as many as 20% of fern species of Struthiopteris belonging to many genera of phylogenetically unrelated families (Watkins et al. 2016). In dimorphic ferns, the two 1a Plants dimorphic or subdimorphic, > 5 cm tall ...... 2 types of fronds differ in morphology in ways associated 1b Plants monomorphic, < 5 cm tall. Iceland ...... S. fallax with their different functions. Typically, the sterile leaf 2a Rhizome scales lanceolate-triangular, concolour, rarely develops with an expanded, well-developed lamina for bicolour; plants subdimorphic, < 20 cm tall ...... photosynthesis, while the fertile type has heavily con- ...... S. spicant var. homophyllum tracted, sporogenous pinnae that lack green tissue when 2b Rhizome scales linear, bicolour; plants dimorphic, usu- mature. Truly trimorphic ferns are much more unusual, ally > 20 cm tall …...... 3 and only a handful of cases have been documented, 3a Plants with just one type of sporophyll ...... mainly falling into the category that Wagner and Wag- ...... S. spicant var. spicant ner (1977) cite as “heterophylly of vegetative leaves”. 3b Plants with two types of sporophylls ...... In these species, besides the presence of sporogenous, ...... S. spicant var. pradae morphologically different fronds, there are two types of sterile ones. The best studied cases are those of some Struthiopteris fallax (Lange) S.Molino, Gabriel y Galán climbing species, which can develop smaller or simpler & Wasowicz, comb. & stat. nov. ≡ Blechnum spicant var. fronds when in contact with soil but produce very much fallax Lange, Fl. Dan. [Oeder] 17(fasc. 50): 11, t. 1988. larger and more divided fronds when in contact with the 1880. ≡ Struthiopteris spicant var. fallax (Lange) Wasow- tree host as for example the Blechnaceae species Icarus icz & Gabriel y Galán, Phytotaxa 302: 198. 2017.—LEC- filiformis (A.Cunn.) Gasper & Salino (Breitwieser et al. TOTYPE: Island, Tunguhver, varme Kilde, Ch. Grønlund 2010–2018). Austroblechnum asperum (Klotzsch) Gasper s.n. (designated in Wasowicz et al. 2017a): C barcode & V.A.O.Dittrich (Blechnaceae) produces long, creep- C10021769, left hand specimen [n.v]). ing, stoloniferous sterile fronds with very reduced lobes (Rodríguez-Ríos 1995) but the contracted fertile fronds Struthiopteris spicant var. pradae S. Molino & Gabriel y are very short, which is intriguing because sporogenous Galán, var. nov.—HOLOTYPE: Spain: Zamora, Aciberos, fronds are expected to be tall in order to help spread- c. 3 km west from the village, in a shadowed slope near the ing the spores. In Platyzoma microphyllum R.Br. (Pteri- road, under a canopy of Fraxinus angustifolia and Quercus daceae) a set of highly reduced, small, filiform fronds are pyrenaica, 42.038942º, −6.861643º, 1220 m a. s.l., 20 Sep produced at the base of the normal ones (Tryon and Tryon 2017, Molino, Seral, de la Fuente and Gabriel y Galán 1982); the function of these, if any, remains unknown. (MACB10654) (Fig. 8). In contrast to these “trimorphic” ferns with two types Diagnosis: This variety differs from the typical var. spicant of sterile leaves, little has been written or observed about in having two types of fertile fronds, one that is larger than ferns with two types of fertile fronds. Wagner and Wag- the sterile fronds and with contracted fertile pinnae, and at ner (1977) stated that the emergence of distinct fertile

1 3 Author's personal copy

The complex Struthiopteris spicant in Western Europe 267 least one that is the same size as the sterile fronds and whose Caherdaniel, 17 Jul 2016, Molino and Pachón (MACB pinnae are not contracted, thus completely resembling sterile 109619). NORWAY. Telemark, Nottoden, 15 Aug 2004, P. fronds, except in being partially fertile. Sunding (MA 747921). SPAIN. Asturias: Valdés, Luarca, 12 Apr 2017, Gabriel y Galán (MACB 109615) (7 individu- Etymology: This variety is dedicated to renowned Spanish als). Canary Islands: Anaga, Tenerife, 25 Oct 1989, San- pteridologist Carmen Prada. tos (MACB 36152); Anaga, Tenerife, 14 Sep 2017, Gabriel Habitats: It usually grows on wet slopes with siliceous soils. y Galán (MACB 110658) (2 individuals). Cantabria: Camaleño, Cosgaya, 8 Oct 2016, Gabriel y Galán (MACB Distribution area: The species is endemic to Spain and 109622) (4 individuals). : Dehesa de Somosierra, occurs in the north of the Iberian Peninsula, in the prov- Madrid 31 May 2017, Gabriel y Galán and Molino (MACB 109611). inces of Zamora (near Puerto de Padornelo), Asturias (near : Aciberos, 23 Sep 2017, Molino et al. (MACB Luarca), Salamanca (Sierra Peña de Francia), and Burgos Zamora 110655). Monte Aloya, Tuy, 15 Jul 1993, (Sierra de la Demanda) (Fig. 9). Pontevedra: Pajarón and Pangua (MACB 59142). UNITED KING- Additional specimens examined: See Appendix 1. DOM. South Somerset, Seven Wells Wood, 30 Jul 1968, J.A. Crabbe Conservation status: Although we do not have enough data (MA 195093). Mid Ebudes, Mull, Tobermory, Crabbe Jermy to propose a specific status yet, we have detected just four 06 Jul 1970, and (MA 195091). populations of this plant and in none of them does the vari- Struthiopteris spicant homophyllum ety seem to be very abundant, so it may be a vulnerable or var. (Merino) Gabriel endangered taxon. y Galán & R.Pino PORTUGAL. Braga, Vieira do Minho, 1 Oct 2004, Acknowledgements The Universidad Complutense de Madrid partially Prada (MACB 109621). SPAIN. La Coruña: Santiago, supported this research through the funding of the research project Cantaleta, 29 Jul 1967, Barrera (MACB 32367). Ponteve- PR26/16-20295 and a field trip to Iceland (International Mobility Pro- dra: between Tabagón y Tomiño, 19 Mar 2016, Gabriel gram 2016). y Galán (MACB 109617); Mondariz, 20 Mar 2016, Gabriel y Galán Compliance with ethical standards (MACB 109618). Salamanca: Batue- cas, 15 May 2016, Gabriel y Galán (MACB 109626) (4 individuals). Conflict of interest The authors declare that they have not conflict of interest. Struthiopteris spicant var. pradae S.Molino & Gabriel y Human and animal rights This research did not involve any testing on Galán humans or animals. SPAIN. Asturias: Valdés, Paladeperre, 22 Mar 2016, Gabriel y Galán (MACB 109613) (3 individuals); 3 Aug 2016, íbidem (MACB 109615) (3 individuals); Appendix 1 Luarca, 26 Jul 2017, Gabriel y Galán (MACB 110660); Otur, 17 Aug 2017, Gabriel y Galán (MACB 110661). Bur- : Sierra de San Millán, 27 Sep 1975, Fuentes (MACB List of materials in alphabetical order of taxa, with indi- gos 5994). : San Miguel de Valero, 14 May 2016, cation of location, voucher, and number of individuals per Salamanca Gabriel y Galán (MACB 109624) (3 individuals). : voucher. Zamora Aciberos, 23 Sep 2017, Molino et al. (MACB 110654). Struthiopteris fallax (Lange) S. Molino, Gabriel y Galán & Wasowicz. ICELAND Deildartunguhver, Wasowicz and Gabriel y References Galán7 Jul 2016, (MACB 109359) (3 individuals). Breitwieser I, Brownsey PJ, Heenan PB, Nelson WA, Wilton AD (eds) Struthiopteris spicant (L.) Weiss var. spicant (2010–2018) Taxon profile—Blechnum filiforme (based on Allan Eber- 1961). Flora of New Zealand Online. Avaliable at: http://www. AUSTRIA. Salzburg, Hohe Tauern, 20 Jul 1994, nzflora.info/facts heet/Taxon /Blech num_filif orme.html . Accessed wein and Vitek (MA 767022). BELGIUM. Turnhout, Engels 1 Aug 2018 Kamp park2, 18 Jun 1973, W. Van Cotthem (MA 809713). Chiou W-L, Shieh W-C, Devol CE (1994) Struthiopteris. In: Huang FRANCE. Sources Foret de l’Esterel, 01 Aug 1958, G. Gav- T-C (ed) Flora of Taiwan, vol I. Editorial committee of the Flora elle P. Bosserdet of Taiwan, Taipei, pp 271–273 (MA 186653). Serans (Oise), 22 Jul 1921, Christ H (1904) Les fougères de la Galicie Espagnole. Bull Acad Int (P 1001059). ICELAND. Vestfirðir, 8 Jul 2016, Gabriel y Géogr Bot 13:76–81 Galán and Wasowizc (MACB 110659). IRELAND. Iveragh,

1 3 Author's personal copy

268 S. Molino et al.

Gabriel y Galán JM, Prada C, Rolleri C, Ainouche A, Vicent M Passarelli L, Gabriel y Galán JM, Prada C, Rolleri CH (2010) (2013) cpDNA supports the identification of the major lineages Spore morphology and ornamentation in the genus Blechnum of American Blechnum (Blechnaceae, Polypodiopsida) established (Blechnaceae, Pteridophyta). Grana 49:243–262. https ://doi. by morphology. Turk J Bot 37:769–777. https ://doi.org/10.3906/ org/10.1080/00173 134.2010.52424 5 bot-1210-49 Perrie LR, Wilson RK, Shepherd LD, Ohlsen DJ, Batty EL, Brown- Gasper AL, Dittrich VAO, Smith AR, Salino A (2016) A classification sey PJ, Bayly MJ (2014) Molecular phylogenetics and generic for Blechnaceae (Polypodiales: Polypodiopsida): new genera, res- taxonomy of Blechnaceae ferns. Taxon 63:745–758. https ://doi. urrected names, and combinations. Phytotaxa 275:191–227. https org/10.12705 /634.13 ://doi.org/10.11646 /phyto taxa.275.3.1 PPG1 (2016) A community-derived classification for extant lyco- Gasper AL, Almeida TE, Dittrich VAO, Smith AR, Salino A (2017) phytes and ferns. J Syst Evol 54:563–603. https://doi.org/10.1111/ Molecular phylogeny of the fern family Blechnaceae (Polypodi- jse.12229 ales) with a revised genus-level treatment. Cladistics 33:429–446. Prada C, Gabriel y Galán JM, Sáiz P, Passarelli L, Ciciarelli MM, https ://doi.org/10.1111/cla.12173 Rolleri C (2016) Caracteres diagnósticos de frondas esporóge- Horjales M, Redondo N, Pérez Prego JM (1990) Nota citotaxonómica nas y esporangios de Blechnum (Blechnaceae). Iheringia Bot sobre pteridoflora del noroeste de la Península Ibérica. Anales 71:161–174 Jard Bot Madrid 48:82–84 Rodríguez-Ríos R (1995) Blechnaceae. In: Marticorena C, Rodríguez Hovenkamp PH (1986) A monograph of the fern genus Pyrrosia. Lei- R (eds) Flora de Chile, vol. 1. Universidad de Concepción, Con- den Bot Ser 9:1–280 cepción, pp 286–303 Iwatsuki K (ed) (1992) Ferns and fern allies of Japan. Heibonsha Pub- Rolleri C, Prada C (2006) Revisión de los grupos de especies del lishers, Tokyo género Blechnum (Blechnaceae-Pteridophyta): el grupo B. penna- Korpelainen H, Pietiläinen M (2008) Effort to reconstruct past popula- marina. Acta Bot Malac 31:7–50 tion history in the fern Blechnum spicant. J Pl Res 121:293–298. Rothfels CJ, Sundue MA, Kuo L-Y, Larsson A, Kato M, Schuettpelz https ://doi.org/10.1007/s1026 5-008-0158-8 E, Pryer KM (2012) A revised family-level classification for eup- Kramer K, Chambers T, Hennipman E (1990) Blechnaceae. In: Kramer olypod II ferns (Polypodiidae: Polypodiales). Taxon 61:515–533 K, Green P (eds) The families and genera of vascular plants: I. Shepherd LD, Perrie LR, Parris BS, Brownsey PJ (2007) A molecular Pteridophytes and Gymnosperms. Springer, Berlin, pp 60–67 phylogeny for the New Zealand Blechnaceae ferns from analy- Lange J (1880) Icones plantarum Florae Danicae Fasciculus L. C Phi- ses of chloroplast trnL-trnF DNA sequences. New Zealand J Bot libert, Copenhagen 45:67–80. https ://doi.org/10.1080/00288 25070 95097 03 Lawalrée A (1964) Blechnum L. In: Tutin TG, Heywoood VH, Burges Soltis PS, Soltis DE (1988) Genetic variation and population structure NA, Valentine DH, Walters SM, Webb DA (eds) Flora Europaea, in the fern Blechnum spicant (Blechnaceae) from Western North vol 1. Lycopodiaceae to Platanaceae. Cambridge University Press, America. Amer J Bot 75:37–44. https://doi.org/10.2307/24439 03 Cambridge, pp 22–23 Tryon AF, Lugardon B (1991) Spores of the pteridophyta. Springer, Löve A, Löve D (1966) The variation of Blechnum spicant. Bot Tidsskr New York 62:186–196 Tryon RM, Tryon AF (1982) Ferns and allied pants with special refer- Löve A, Löve D (1968) Cytotaxonomy of Blechnum spicant. Collect ence to tropical America. Springer, New York Bot 7:665–676 Vicent M (2017) Sistemática de la familia Blechnaceae (Polypodiop- Lugardon B (1965) Strutcture des parois de la spore de Blechnum spi- sida): filogenia de Lomaridium y Parablechnum p.p. y aspectos cant (L.) Roth. Pollen Spores 7:409–428 evolutivos de las pinnas fértiles. PhD Thesis, Universidad Com- Lugardon B (1974) La structure fine de l’exospore et de la périspore plutense, Madrid des Filicinées isosporées, II. Filicales. Comment. Pollen Spores Wagner WHJ, Wagner FS (1977) Fertile–sterile leaf dimorphy in ferns. 16:161–226 Gard Bull Singapore 30:251–267 Merino B (1898) Contribución a la Flora de Galicia, Suplemento I. Wasowicz P, Folcik L, Rostanski A (2017a) Typification of Blechnum Tipografia Regional, Tuy spicant var. fallax Lange (Blechnaceae). Acta Soc Bot Poloniae Molino S, de la Fuente P, Seral A, Gabriel y Galan JM (2018) Nuevas 86:3542. https ://doi.org/10.5586/asbp.3542 localidades de tres helechos para la Península Ibérica. Bot Com- Wasowicz P, Gabriel y Galán JM, Pino R (2017b) New combinations in plut 43:2–6 Struthiopteris spicant for the European flora. Phytotaxa 302:198– Nakato N (1987) Chromosome numbers of three endemic species of the 200. https ://doi.org/10.11646 /phyto taxa.302.2.11 fern genus Blechnum in Japan. J Jap Bot 62:129–133 Watkins JE, Churchill AC, Holbrook NM (2016) A site for sori: eco- Nauman CE (1993) Blechnum. In: FNA Editorial Committee (ed) Flora physiology of fertile-sterile leaf dimorphy in ferns. Amer J Bot of North America North of Mexico, vol. 2. Oxford University 103:845–855. https ://doi.org/10.3732/ajb.15005 05 Press, New York, Oxford, pp 223–227 Ormonde J (1986) Blechnum. In: Castroviejo S, Laínz M, López Publisher’s Note Springer Nature remains neutral with regard to González G et al (eds) Flora Iberica I, vol. 1. CSIC, Madrid, pp jurisdictional claims in published maps and institutional affiliations. 151–153 Passarelli L (2007) Estudios esporales en especies del grupo Blech- num penna-marina (Blechnaceae-Pteridophyta). Acta Bot Malacit 32:1–19

1 3